OECD AI Adoption Trajectory 2020–2024
Headline figures
OECD-wide share of firms with 10+ employees using AI:
| Year | Share |
|---|---|
| 2020 | 5.6% |
| 2021 | 7.3% |
| 2022 | 8.1% |
| 2023 | 8.5% |
| 2024 | 14.0% |
OECD-wide AI adoption roughly 2.5× over five years; the largest year-on-year jump (8.5% → 14%) is 2023→2024, coinciding with the generative-AI inflection.
Source figure: oecd-sme-ai-adoption-2025, Figure 1, p. 9 (OECD ICT Access and Usage Database, accessed July 2025).
Comparison to other digital technologies
| Technology | 2024 OECD adoption |
|---|---|
| Cloud computing | >50% |
| AI | 14% |
AI is well below the diffusion level of mature digital technologies. It is still on the steep early part of the S-curve; cloud computing is past its steepest part.
Caveats
- Unweighted average across OECD Member countries; country composition varies year to year as data become available.
- Definitions of “AI use” have expanded over the period — 2024 survey instruments capture generative AI explicitly, which earlier instruments do not. Some of the 2023→2024 jump is real adoption acceleration; some is definition broadening.
- Cross-country comparability caveats apply throughout.
Why this matters
- The trajectory is real but the level is low. AI is not (yet) a generalized firm-level technology. At 14% with the gap widening by firm size, the OECD economy is still on the early phase of the diffusion curve.
- The gap is widening in absolute terms. Even as adoption grows, the gap between small-firm and large-firm adoption rates is increasing in percentage-point terms — a structural divergence, not a temporary lag.
- Per diffusion-vs-innovation, the trajectory is the load-bearing metric. If the curve continues to flatten at the SME tail, the diffusion-deficit hypothesis hardens.
Related
- oecd-sme-ai-adoption-2025 — primary source.
- sme-ai-adoption-gap — the gap that grows with the level.
- sme-vs-large-firm-ai-gap — the by-size decomposition.
- oecd-sectoral-ai-adoption-2024 — the by-sector decomposition.
- diffusion-vs-innovation — diffusion S-curve framing.
- H2_u-curve-of-value — the application layer’s broad base measured here.
- global-ai-spend-632b — the investment side of the same story.